Close-the-Loop Follow-Up Tracker

The problem this kills

You run a real close-the-loop program. Every survey cycle, a pile of detractors and pointed comments lands in your lap and you swear you'll follow up on all of them. Then reality hits. You drop the flagged items into a spreadsheet, hand-assign owners in a column, and email a few people to say "can you reach out to this customer?" A week later you genuinely cannot tell which ones got handled. Did Maria call the account that gave you a 2? Did anyone ever respond to the customer who wrote "I've been a customer for six years and nobody returned my call"? You ask in Slack, get three "I think so"s, and quietly mark a few rows green so the report looks finished.

That's the trap of close-the-loop done by hand: items get assigned but never chased, owners close cases with no record of what was actually said to the customer, the same piece of feedback gets followed up twice by two different people, and your headline metric - how fast you actually close the loop - is a number you can't defend because the underlying data is a colored-in guess. When the VP asks "what's our loop closure rate and how long does it take us?", the honest answer is "let me get back to you."

This tool turns close-the-loop from a hopeful spreadsheet into a clean, accountable process: every flagged item becomes an owned task with a due date, overdue ones get nudged automatically, the owner has to record what happened before anything can close, and a manager signs off on "loop closed." Nothing slips, nothing closes empty, and your closure-rate and time-to-close numbers are real.

What you'll build

A small internal web app, just for your CX team, that:

• Loads your flagged feedback from a CSV - detractors, low CSAT, and any actionable comment - with the score/type, the customer, and the owner.
• Creates a follow-up task per item with an owner and a due date computed from your follow-up window.
• Shows a live open-loops board: who owns what, when it's due, and whether the loop is actually closed - color-coded by on-time / due-soon / overdue.
• Nudges overdue owners automatically via Resend (and optionally their manager), so nothing ages out silently.
• Makes the owner record the follow-up outcome - what they did, what the customer said, what was promised - before a case can move forward.
• Requires a manager to approve "loop closed" as a hard human gate; a case cannot be marked complete without a recorded outcome.
• Dedupes on feedback item ID so the same flagged item can never become two tasks.
• Reports your closure rate and the headline time-to-loop-closure, and exports a clean open-loops worklist as CSV.

What's inside the Implementation Plan

The plan is a single markdown file you paste into Claude Code (a free AI coding agent). It walks the agent through building the whole tool, step by step, each step ending with a ready-to-paste prompt.

The most important part: the plan opens by interviewing you about your business - so you get a tool shaped to your VOC program, not a generic template you have to bend into place. Before it writes a line of code, the agent asks how flagged feedback reaches you today, what survey platform it comes from (NPS, CSAT, a Sheet), the real column names and how you decide an item is "actionable," how owners are assigned, your follow-up due window and any priority tiers, exactly who has to approve a closure, and your messiest edge cases - anonymous responses, the same customer flagged twice in a cycle, items nobody owns. It reads a short tailored spec back to you, you confirm it, and only then does it build.

Inside you'll find:

• The discovery interview and how the agent turns your answers into the data model, the due-date rules, and the closure-approval flow.
• The full build: database, login, CSV import with a duplicate guard on feedback item ID, the task/owner/due-date engine, the open-loops board, the overdue nudges via Resend, the outcome-logging step, the manager closure gate, and the metrics.
• The hard human approval gate, the "no closure without a recorded outcome" rule, and the duplicate guard.
• Verification steps so you can prove it works, and the CSV-export fallback so the tool is fully usable today even before you connect it to your survey platform.

The governance it includes (this is the point)

This isn't a toy. The plan builds in the controls a CX program actually needs to be trusted:

• Login so only your team can see or touch anything.
• Row-level security so people only ever see the feedback, tasks, and outcomes that belong to your organization.
• A complete audit trail - every import, assignment, nudge, outcome logged, and approval recorded with who and when.
• A hard human-in-the-loop gate - the tool drafts and tracks, but a manager must approve "loop closed" before a case is complete, and no case can close without a recorded follow-up outcome.
• Duplicate guards - dedupe on feedback item ID so the same flagged item can never be tracked or closed twice.

Who it's for

CX managers, VOC program leads, customer success ops, and support team leads running a formal close-the-loop program who today track detractor follow-up in a spreadsheet and can't honestly report their closure rate. If you're the person who has to answer "did we follow up, and how fast?" - this is for you. You don't need to write code. You need your flagged-feedback CSV, your follow-up rules, and an afternoon.

You've got this - paste the first prompt and let the agent interview you.